Tuesday, 24 November 2015

Why are there Petrine writings?

At SBL this weekend I picked the excellent 2015 volume from Eerdmans, Peter in Early Christianity, edited by the incomparable Helen Bond and Larry Hurtado. There is in this book a quite interesting contribution from Matthew Novenson, entitled "Why are there some Petrine epistles rather than none?" In it Novenson operates from two suppositions: 1) the four known Petrine epistles (1 and 2 Peter, the Letter of Peter to Philip, and the Kerygmata Petrou) are pseudepigraphic, and 2) pseudepigraphical epistles are quite rare, compared to pseudepigraphical gospels and apocalypses. These two suppositions lead him to ask, Why did this relatively rare corpus of pseudepigraphical originate? His answer is that someone wrote 1 Peter in the name of the apostle, and then the authors of our other three epistles followed that person's lead. Although quite possible, this seems to some extent to leave the question unanswered, as it does not clearly explain why 1 Peter was written in the first place. At the end of the chapter we still do not know why there are some rather than none.

Perhaps Novenson makes it too hard on himself. If he were to judge that 1 Peter was actually written by Peter, allowing for the fact that the realities of ancient authorship do not require the apostle to have actually penned the words in the letter (thus removing considerations such as his levels of literacy and education), then in fact Novenson has a ready-made answer: Peter wrote an epistle and this inspired a tradition of emulators. Why are there some Petrine epistles rather than none? Because Peter wrote at least one epistle. All that would then be left to do is the mundane historical work of defining the situation that led Peter to pen 1 Peter.

So, then, let us consider: what are Novenson's arguments that Peter did not write 1 Peter, and do they actually warrant that judgment? It turns out that the arguments are only derivatively against Petrine origin, and more precisely against a pre-70 date. This is itself a debatable move, for if one is quite prepared to dispense with the traditions that connect Peter with 1 Peter then can one be quite so confident in the traditions that indicate that Peter died before 70? Surely both must be up for debate, and if so then it is not self-evident that a post-70 date would exclude Petrine authorship. But, given my vaguely neurotic obsession with the dates of the New Testament texts, I find myself focusing upon the argument for a post-70 date for 1 Peter, and find myself less-than-entirely persuaded by the arguments advanced in the chapter in order to establish the case. (In what follows I will quote from Novenson, then offer my rejoinder).

1)     “[T]he use of the name of Babylon for Rome.” The initial, negative rejoinder to be made is that this is entirely an argument from silence and thus of limited probative value; after all, we do not have positive evidence that Rome was not referred to as Babylon before 70. The argument risks further begging the question: the judgment can only be sustained if we have antecedently judged 1 Peter to have been written post-70, because otherwise it would constitute positive evidence of such usage pre-70; yet that is exactly the question at hand. Positively, we might well have evidence of such usage pre-70, as there is a respectable minority in the New Testament guild that dates Revelation to before 70. This is a reminder of John A.T. Robinson’s insight that one cannot consider the dates of the NT (and other first-century Christian) texts in isolation from but rather in relation to one another: one’s judgment on the matter of Revelation’s date will potentially impact one’s judgment regarding that of 1 Peter, for if Revelation is written in the late-60s then one could hardly rule out that Peter, who died just a few years earlier on general consensus, was unfamiliar with the term. Further, both 1 Peter and Revelation seem to assume that the readers are familiar with the usage, such that even if we date Revelation to the 90s we have not evidence that the term was in use no earlier than that point but rather no later. Considering all the above we simply do not have a terminus post quem for the usage, and thus to use it to establish such a terminus is a debatable move.

2)     “[T]he use of the name Christians for the recipients.” Again, the supposition is that “Christian” only emerged post-70, and again, this is entirely an argument from silence that risks begging the question. It is again empirically questionable, as the Acts of the Apostles tells us that the movement was already called “Christians” quite early, situating the story to which Luke connects the origin of the name no later than the end of the 30s. Again, other judgments about dating become relevant: if, as Harnack came to hold at the end of his life and a minority camp has always argued, Acts was written around 63 or 64 then we have evidence that this term was used to describe members of the movement before 70. Likewise, given that most scholars would date the Acts to the 80s and 90s, and that it seems to suppose that the usage is already known to the readers, it seems to provide no terminus post quem and is thus non-probative in this instance.

3)     “[T]he presence of established churches throughout Asia Minor.” I must admit that I find this to be a somewhat surprising argument, as it must dispense with a wealth of evidence, such as the core of the Acts narrative (which has Paul founding churches throughout Asia Minor in the 40s and 50s) and Pauline or pseudo-Pauline texts such as Galatians (written to the churches of Galatia), Colossians (which not only is addressed to Colossae but also mentions churches in Laodicea and Hierapolis, all of which are Asian cities) and 1 Timothy (which has Timothy in Ephesus). Granted outstanding questions regarding the accuracy of the Acts narrative and the Pauline authorship of certain of these particular texts there nonetheless seems a clear Christian memory that no later than c. 50 there were Christians in Asia Minor, and their extent throughout the region is far from clear. Certainly the evidence does not readily support the negative judgment that there not churches throughout Asia Minor, and indeed unless one wants to call into question the authenticity of Galatians or the date of Paul’s death (roughly coeval with that of Peter, on the consensus date, which, without further argument, we ought to suppose that Novenson assumes given his subscription to the closely-related consensus date of Peter’s death) then we know that there were churches in at least Galatia (one of the locales mentioned in 1 Peter 1:1) during Peter’s lifetime. An argument from silence in which there is no silence at all should probably be judged to be a less-than-compelling position.

4)     “Among other reasons.” Remaining uncited and thus inaccessible to critique such reasons do not exist in Novenson’s argument. We can thus dispense with them.

The point of the above is not to argue that 1 Peter is pre-70 or that it was written by Peter; such positions would require a great deal more work than this. The argument is that the supposition that 1 Peter is pseudepigraphical requires more than showing it to be post-70 (as one must also show that Peter died pre-70 for that to serve as an argument in relation to Petrine authorship), and the judgment that the text is post-70 requires more work than is frequently undertaken in contemporary scholarship. Too often we simply assume that we know which NT texts are genuine and which are not: Romans and the other six undisputed are immune from arguments for inauthenticity; Colossians, Ephesians, and 2 Thessalonians remain always a coin flip (and thus are usually treated functionally as pseudepigraphical, lest one be vulnerable to the objection “I just don’t think Paul wrote that”); the Pastoral Paulines are immune from arguments for authenticity; and the Catholic epistles tend to be treated functionally as pseudepigraphy. Yet most of these judgments were reached over a century ago, and we’ve learned a great deal more about the ancient world since then. We need to be able to render good accounts of why we still affirm these judgments, or else perhaps we don’t actually know what we think that we know.